Fraunhofer Institute for Cognitive Systems IKSFrom October 18 to 23, 2022, the IEEE Quantum Week will take place in Broomfield, Colorado, USA.
The event is bridging the gap between the science of quantum computing and the development of an industry surrounding it.
The Fraunhofer Institute for Cognitive Systems IKS will be present with two talks:
Quantum-classical convolutional neural networks in radiological image classification
On Monday, 19th September, Maureen Monnet, scientist at Fraunhofer IKS, will talk about "Quantum-classical convolutional neural networks in radiological image classification".
Accurate and reliable predictions are crucial in medical imaging diagnostics. While clinical studies typically achieve sample sizes of around 100 to 1000, this is often not sufficient for ML approaches. On the other hand, quantum computing assisted algorithms promise to achieve high prediction accuracy even with limited amounts of data. The authors propose and analyze the performance of different hybrid quantum-classical convolutional neural networks (QCCNN) on 2D and 3D medical imaging data to identify malign lesions.
Quantum Robustness Verification : A Hybrid Quantum Classical Neural Network Certification Algorithm
Nicola Franco, scientist at Fraunhofer IKS, will give a presentation about "Quantum Robustness Verification : A Hybrid Quantum Classical Neural Network Certification Algorithm" on Tuesday, 20th September.
In this work, the authors propose to use quantum computing for neural network verification, which involves solving many-variable mixed-integer programs (MIPs). They further improve existing hybrid methods based on the Benders decomposition by reducing the overall number of iterations and placing a limit on the maximum number of qubits required.